Some chronic conditions require patient self-management and behavior modification based on the regular collection of health status data from monitoring devices. Given the high complexity of potential solutions, we focus on a particular use case. Because we will adopt portable, reusable, and extensible ontology-based knowledge representation and reasoning techniques we expect to build foundations for the future development of patient-centered decision aids. To demonstrate the proposed disease self-management decision aid we will implement and evaluate a patient- centered decision system for adult type 1 diabetes outpatients who are using insulin pumps to support the daily task of adjusting pre-meal insulin dosage. As the abundant literature about noncompliance indicates, predetermined self-management care programs are not effective in diabetes care, as they are not tailored to patient priorities, goals, psychosocial factors, and lifestyle. Our hypothesis is that incorporating patient preferences, glycemic and well-being goals, and adherence to health-behavior recommendations will improve decision-making and postprandial glucose levels for individuals with T1D. Our first aim is to learn about diabetes patient preferences and goals by interviewing patients and through bibliographic surveys. Our second aim is to capture in an ontology the knowledge gained from first aim, and build on top of it an argumentation-logic decision aid that suggests treatments that are tailored to patient's clinical state and preferences, in order to achieve active clinical goals. When more than one treatment is available to achieve the clinical goals, the argumentation system compares and ranks options by presenting human-like arguments for and against the proposed treatments. Our third aim is to deploy and evaluate the decision aid. I DECIDE will be deployed as a smartphone application. It will ask patients at the time of the pre-meal insulin dosage their preferences on carbohydrate and alcohol intake, planned after-meal exercise, and provide them suggestions to help them achieve tailored clinical goals. Recommendations will be supported with lay-based explanations to help patients learn how daily choices affect their overall well-being and goals achievement. The decision system will be evaluated with a feasibility study involving adult T1D outpatients in terms of evidence-based compliance, accuracy and comprehension.